An official website of the United States government.

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Assuring Fruit and Vegetable Product Quality and Safety Through the Handling and Marketing Chain

Delwiche, Michael
University of California - Davis
Start date
End date
Develop, evaluate, and apply rapid sensing technologies to assure food safety including bio-security, purity, and integrity of produce.
More information
NON-TECHNICAL SUMMARY: The development of sensing systems for detection of pathogenic bacteria on fruits and vegetables will lead to improved food safety.

APPROACH: Sensors to detect bacterial pathogens in the irrigation and wash water of fresh and minimally processed fruits and vegetables will be designed. The first application will be to detect Salmonella in alfalfa sprout irrigation water. A real-time PCR has been developed and will be adapted to work in the sensor. Thermal cycling of the reaction cell will be done with a thermoelectric heater/cooler and embedded controller. Amplified DNA fragments will be measured using fluorescent markers and solid-state optical sensors. An automated system to sample the water and concentrate the cells will be developed. The entire system will be mounted in a small waterproof enclosure and tested with several serovars in distilled water and sprout irrigation water to determine sensitivity and specificity. The potential for detection of other pathogens will be evaluated.

PROGRESS: 2002/10 TO 2007/09
A nucleic acid sensor capable of automatically loading PCR reagents and a test sample, running real-time PCR, and automatically cleaning the sample carrying line was built and tested. The system consisted of an integrated thermal cycler, fluorometer, and fluid handling components. An aluminum reaction chamber housing was designed to transfer heat, secure contact between the glass reaction chamber and thermistor for temperature feedback control, provide fluorescence excitation and emission pathways through fiber optic cable holes, and facilitate opening and sealing of the reaction chamber on the thermoelectric module for automated fluid handling. Fiber optic cables were connected to the aluminum reaction chamber at right angles for transferring excitation and emission light signals. A high gain photovoltaic amplifier circuit was designed and allowed for the amplification of photodiode signals in response to emitted fluorescence light. The system displayed both temperature and cycle number during the PCR. Only four commands on the embedded controller were required to load the reagents, run the reaction, clean the sample carrying line, and download the results. Taq polymerase concentrations were five times higher in these experiments as compared to PCR reactions run previously with a commercial instrument. A high annealing temperature of 62 C (primer Tm = 54 C) was also used for the real-time PCR experimentation with the nucleic acid sensor. The total time to mix and load the test sample and reagents, run the reaction, and clean the sample carrying line was 2 hours and 20 minutes, significantly less time than would be required to have a sample shipped and tested by an independent laboratory. Even with high concentrations (50 ng/rxn) of Salmonella enterica Newport DNA, it was possible to clean the sample carrying line and tube for reuse without noticeable carry-over contamination between alternating positive and negative control reactions. Equally important, the reagents used to clean the sample carrying line did not appear to inhibit PCR, as the amplification fluorescence appeared to increase at the same point for each control reaction, and the magnitude of each dissociation peak remained constant. The nucleic acid sensor system gave repeatable results with both positive and negative control reactions as seen with the amplification and dissociation data. The nucleic acid sensor was calibrated in autoclaved and 48-hour sprout irrigation water. S. enterica boiled cells were detected over a range of approximately 104 to 108 CFU/rxn. It was possible to generate enough PCR product to visualize a band on a gel at the expected size over approximately five orders of magnitude from 3.2 x 103 to 108 CFU/rxn. Automated detection experiments yielded correct identification of 9/9 positive control reactions over a range of 104 to 108 CFU per reaction and 1/1 negative control reactions. Primer dimers were not seen in positive or negative control reactions with sprout irrigation water, suggesting that it may be possible to improve the detection limit by increasing the number of thermal cycles or by lowering the annealing temperature.

IMPACT: 2002/10 TO 2007/09
Recent outbreaks of Escherichia coli O157:H7 on spinach and lettuce have increased concerns about the possibility of pathogenic bacteria entering the food chain by way of minimally processed produce. Several outbreaks of hazardous bacteria have also been linked to sprouts. The process of growing sprouts, such as alfalfa and bean sprouts, can permit the growth of bacteria due to warm and moist conditions found in many sprouting operations. As a result, sprout growers are required to test sprout water samples for Salmonella and E. coli O157:H7. Testing may be done by shipping samples to a fully equipped microbiology laboratory for analysis. However, shipping water samples off-site is a time consuming process. On-site testing with an automated sensor would let sprout growers know immediately if their product was safe to ship. While a variety of commercially available real-time PCR instruments exist, they are typically large and require skilled labor to operate and extensive equipment for liquid handling. Screening samples for the same bacteria in the same media (sprout water) is a repetitive task that is well suited to automation. We have developed a real-time PCR system that simplifies testing sprout water for pathogenic bacteria with PCR by automatically loading a sample and reagents, running the reaction, interpreting the results, and cleaning the sample line. Automated sensor systems that facilitate the process of testing produce for pathogens can minimize the potential for harmful bacteria to reach the consumer. The results indicate that with continued development, it is feasible to use a real-time PCR system for the rapid on-site screening of water for pathogens without the need for a fully equipped laboratory and without the need for a technical staff to operate equipment and interpret results.

Funding Source
Nat'l. Inst. of Food and Agriculture
Project source
View this project
Project number
Accession number